OBJECTIVES: (i) To investigate the effects of current obesity, steady weight change and weight¯uctuation on physical functioning and (ii) to determine whether associations are independent of coronary heart disease. DESIGN: Prospective cohort study with body mass index (BMI) measurements at four time points between the ages of 25 and 63 y. SETTING: British civil servants based in London of®ces at baseline. PARTICIPANTS: 6895 men and 3413 women aged 35±55 y at baseline. MAIN OUTCOME MEASURE: Physical functioning was assessed using the 10-item scale from the Short Form 36 Health Survey, with a score in the lowest quartile indicating poor physical functioning. RESULTS: After adjustment for age and confounders (employment grade, smoking, alcohol, exercise and menopausal status), current BMI was monotonically associated with poor physical functioning in women whereas a threshold effect at a BMI of 27 kgam 2 was seen in men. The odds ratio of poor physical functioning was 1.55 (95% con®dence interval (CI) 1.02±2.35) amongst women and 1.04 (95% CI 0.77±1.41) amongst men with BMI 23±24.9 kgam 2 compared to those with BMI`21 kgam 2 . Women in the upper, compared to the lower, tertile of steady weight change, had an odds ratio of poor physical functioning of 1.79 (1.24±2.60) after adjustment for age, confounders, current BMI and weight¯uctuation. Women in the upper, compared to the lower, tertile of weight¯uctuation had an odds ratio of poor physical functioning of 1.70 (1.23±2.34) adjusting for age, confounders, current BMI and steady weight change. Adjustment for the presence of coronary heart disease did not substantially alter any of these associations. Steady weight change and weight¯uctuation had no independent effects in men. CONCLUSION: Among women, current obesity, steady weight change and weight¯uctuation are independently and monotonically associated with poor physical functioning. Development of overt coronary heart disease is unlikely to be the mechanism for these associations.